That the MCAS may have an issue needing to be addressed isn't in dispute but you haven't proved the MAX is unstable in flight. You are assuming that based on as yet incomplete data about the circumstances of the ill-fated flight and your own insistence that the LEAP engines greater weight and forward placement alter the plane's CoG so much it's fundamentally unstable.
Some of this discussion about stable / unstable is due to improper use of the English word "stable / unstable".
Put an egg on a table, lying on the side, rock the table, and the egg will sway a little from side to side. But as soon as you don't touch the table any longer, the egg will stop swaying. It is stable.
Put the egg on its end. Theoretically you may make it stand on the end, but as soon as you rock the table it will turn 90 degrees to its flatter side. On its end It is unstable.
Now put also a matchbox on the table, and rock the table. The egg will sway a little, while the matchbox will stay as glued to the table. The matchbox is stable. But is the egg now unstable?
No, the egg is as stable as before, but the matchbox has a larger stability margin
. It will take more severe rocking of the table to make the matchbox turn over its end and fall on the floor than needed for the egg to do so.
All transport aircraft are stable on the pitch axis, also FBW. All aircraft, which are unstable on pitch, are military planes, and always FBW because the human brain isn't fast enough to control unstable flight. Their failure mode then isn't "alternate mode", but rather "ejection seat".
All transport aircraft must, to be certified, have a certain minimum level of stability margin.
In one certain corner of its flight envelope the 737MAX has a stability margin, which is lower than wanted, lower than the 737NG (commonality), or maybe even lower than demanded for certification. That is fully corrected by the MCAS system. It makes the MAX act exactly as if it was an NG.
The minimum stability margin must be maintained with CG all way from its forward to its backward limit. There is little doubt that a slight move forward of the back CG limit would have done the same trick. But it was hardly good PR to limit the CG range related to the NG.
The problem is that the MCAS software is another system which can fail. In the Lion Air case it failed because it was fed faulty input data. The crew could have saved the flight, but they failed to do so. Maybe because they didn't know about MCAS, maybe due to some other confusion. We will know more when we see the final report and/or the CVR data.
At the end of the day, for me it is a surprise that a plane, which is certified in the 21st century, has input to automatic primary flight control which is dependent on a single sensor, and "manual backup" with a second sensor, and not triple sensors with voting on agreement. Or at least automatic disconnection of automatic systems when the two sensors disagree. It seems rather "19th century" to me. I have seen plenty of examples of much better design throughout my 20th century lifeline.